Filter for RF noise on AM radio stations?

Any specifics on your desktop? Different CPU's use different power levels. The 80 watts and up variety tend to be rather noisy. Might as well disclose the maker and model number of your AM radio.

1. Attach a big external antenna to your radio. The weak signal will now be stronger, which causes the receive AGC to reduce the receiver sensitivity, which means you'll hear less noise.
2. With the computer running, unplug every cable and contrivance except the power cord. Turn OFF all network appliances, network printers, scanners, fax machines, or electronic devices that might have a clock, microprocessor, or noise generator inside. Do any of them make a big difference in the RFI level? If so, attach a big ferrite bead or toroid to the cable, running as many turns of cable through the bead as possible. More than one bead on a cable is better. If not, keep turning things off until the noise goes away.
3. Also put a ferrite bead on the power cable.

1. USTP CAT5 Ethernet cables tend to generate quite a bit of RF. Much depends on the source of the network noise. Most of the local hams that are into contesting or weak signals have big ferrite beads on their network cables.

2. If mouse movement changes what you're hearing, it's unlikely to be the switching power supply. However, if you happen to have a spare power supply, try it and see if it helps. PC power supplies are rather noisy and operate around 33 KHz. You might be hearing the harmonics. However, the computer power supply is never really turned off until the cord is unplugged. (That's for WOL or wake on LAN). If the noise goes away ONLY when you pull the plug, replace the power supply.

1. Grab a roll of aluminum foil and temporarily wrap the machine in foil. Never mind ventilation or access to controls and connectors for now. Just wrap it and see if shielding helps. If it does, remove the foil, and try shielding various parts of the computah. For something more permanent, aluminum foil duct tape.

2. Google for what ham radio operators do to reduce computer noise.

1. Give up and switch to FM radio. Much nicer, no noise, and generally better programming.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558

The noise is coming from the CPU supply which is always ON, just like a VCR or a TV; the extra +5 line is used for the "power on" logic...

I'm not sure how useful some of these suggestions will be considering that the noise is present even with the computer turned off as long as it is plugged in. This is almost certainly because of the PSU rather than any of the CPU related circuitry. The PSU supplies power to a small portion of the computer so it can turn itself back on. I bet a ferrite on the power cord would help.

--

Rick

So, why does moving the mouse change the sound when the computah is running? The power supply does not change switching frequency with mouse movements. There's also no load change with mouse movements. In my never humble opinion, there are multiple sources of EMI/RFI, one of which is the +5VSB section in the power supply and the other is some kind of CPU related junk being radiated or conducted via some wire or ungrounded piece of sheet metal on the case.

For fairly low frequencies, such as 600 KHz, there has do be some fairly long wires involved in playing antenna. Maybe the power supply wires or some of the other spaghetti wires inside the box. (Wires radiate. Components do not).

So, after the OP is done testing a replacement power supply, which hopefully will fix the +5VSB problem, but finds himself with some remaining junk coming from elsewhere, perhaps the rest of my suggestions might be worth trying.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558

I think you idea of using an external antenna might serve best for DX on 600 kHz. Coming up with a proper interface might prove difficult.

It could be a wifi or otherwise wireless mouse.

joe

** Most SMPSs generate at least a little AM band EMI, learn to live with it by not putting your radio right next to one.

Try putting the radio well away from the PC and orient it for best reception. Run a twin lead from the radio's speaker output to a speaker next to the PC.

.... Phil

I've got a pc here that does the same. In my case its the LAN port that is the source of noise. Totally obliterates AM through a brick wall. I tried all sorts of ferrites etc. My solution was to turn the PC off at the wall.

On Sun, 20 Sep 2015 19:35:27 +0800, Oltimer Gave us:

Go into the motherboard settings and turn OFF the "wake on LAN" setting.

Don't confuse sources and transmission paths. Of course there *can* be multiple sources, but it is unlikely the mouse is one of them. But the mouse can be conducting RFI out of the unit and acting as an antenna.

Like mouse or power cables? Hard to get rid of them unless he goes wireless on the mouse and... adds a power cable with a nice fat RF choke built in.

--

Rick

Thanks, I'll try that.

What kind of display? Sounds like a CRT!

Yep, except the OP indicated that "It's so bad, I can sometimes hear changes in the noise as I move the mouse pointer around on the screen." I interpret that to mean that noise changed in it characteristic frequency components, and not just a change in level. If moving the mouse cord changed only the transmission characteristics, there would only be a change in noise level. If moving the mouse changed something on the screen, which would change the frequency characteristics of something like the video card, then I would guess(tm) that he was hearing something coming from the CPU or video card.

Yep. However, there are plenty of other computah cords and cables that are capable of belching RF. Add the USB, video, PS/2 keyboard, PS/2 mouse, speaker, microphone, USB camera, external USB drives, etc. Also, don't forget about the wall warts that power external computah related devices, such as a modem, router, switch, IP phone, etc.

I have an Alinco DJ-X2 scanner/receiver that I use for RF noise "sniffing". I tune it a likely frequency, set it to AM receiver, and probe around the computer area looking for RF trash. Switcher type wall warts seem to be the worst. Computahs with high power CPU's and video cards are another. Laser printers also seem to be quite noisy when running. Plasma TV's should be banned. A few CFL and LED lamps are rather noisy. I'm sometimes surprised at the source, such as dehumdifiers, negative ion generators, electric power tool chargers, door bells, security systems, and anything with a commutator type motor. I also carry a small selection of clamp on ferrite beads and toroids for the inevitable quick fix.

Most houses have multiple sources of EMI/RFI. One of my desperation tricks is to turn off the entire house. I switch off the power at the breaker panel and remove the batteries from anything running on battery. Then, I listen with the owners receiver to see if there's anything left. That's usually when I find that the neighbors are generating as much or more noise than the home owner. Conducted radiation from power line connected devices carry quite a long distance when using the power lines as a transmission line and/or antenna.

Basically, there are two ways to do this. One is to leave everything on and turn off devices until the noise goes away. That seems to be the approach used by the OP in turning off and unplugging the computah. The other is to turn everything off, and turn them back on one at a time until the source of the noise appears. I prefer the latter because it also includes neighborhood noise sources.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558

It's a small LCD 12 inch I think. There are no other devices connected to the computer other than a wired mouse, the display, keyboard, and a phone line plug. I have another old PC 900MHz running HP-Home that doesn't cause any interference. The newer PC is a 1.6 gig running HP professional that causes all the problems. I think I'll try the ferrite core idea on the power cable. I have some fairly large cores laying around that might help.

--- news://freenews.netfront.net/ - complaints: snipped-for-privacy@netfront.net ---

Possible, but you are making assumptions about frequency changes. Simple changes in relative amplitude can cause the same change in sounds.

--

Rick

Moving the mouse will wake-up the CPU, suddenly switching on a 50-100 Watt load

Nope. That's true as the CPU transistions through various "C" style power saving modes: However, EMI/RFI tests are normally not performed when the computah is in standby or asleep. If that's how you use a computah, you might as well unplug it and listen to the radio station without the added noise.

In Windoze, you can see for yourself by firing up the Task Manager (right click on the task bar and select Task Manager). On the Performance tab will be a graph of CPU Usage History, which should give you an idea of how much CPU horsepower is used to service the mouse. I'm doing it as I type with Windoze XP and Win 7 and find little change in CPU activity with mouse movement. At "idle", the CPU Usage is zero. When I furiously move the mouse around the screen, it creeps up to maybe 7%.

You can see it better by running the Performance Monitor: start -> run -> perfmon.msc The %Processor Usage stays near zero when moving the mouse.

If you add the graph: Add Counter -> ProcessorPerformance -> Power Consumption for each core, I still don't see much change with mouse movement.

Same thing when I display: Add Counter -> ProcessorPerformance -> % of Maximum Frequency

It takes so little CPU horsepower to service the mouse, that the CPU can do it in its sleep.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558

That proves nothing since it is an average of some time period.

It seems completely possible that moving the mouse causes an irq on each mouse "step", say. The CPU wakes up for a microsecond then goes back to sleep (or reduces its power if you prefer). That could cause spikes of

10s of amps at a kHz or so, generating EMI. And in fact I have heard something very like that as interference from computer sound systems, when the mouse is moved with the volume turned right up. A sort of whine with frequency proportional to mouse speed.

--

John Devereux

+1 Mark